Search

Background: Silver-based products have been marketed as an alternative to antibiotics, and their consumption has increased. Bacteria may, however, develop resistance to silver., Aim: To study the presence of genes encoding silver resistance (silE, silP, silS) over time in three clinically important Enterobacteriaceae genera., Methods: Using polymerase chain reaction (PCR), 752 bloodstream isolates from the years 1990-2010 were investigated. Age, gender, and ward of patients were registered, and the susceptibility to antibiotics and silver nitrate was tested. Clonality and single nucleotide polymorphism were assessed with repetitive element sequence-based PCR, multi-locus sequence typing, and whole-genome sequencing., Findings: Genes encoding silver resistance were detected most frequently in Enterobacter spp. (48%), followed by Klebsiella spp. (41%) and Escherichia coli 4%. Phenotypical resistance to silver nitrate was found in Enterobacter (13%) and Klebsiella (3%) isolates. The lowest carriage rate of sil genes was observed in blood isolates from the neonatology ward (24%), and the highest in blood isolates from the oncology/haematology wards (66%). Presence of sil genes was observed in international high-risk clones. Sequences of the sil and pco clusters indicated that a single mutational event in the silS gene could have caused the phenotypic resistance., Conclusion: Despite a restricted consumption of silver-based products in Swedish health care, silver resistance genes are widely represented in clinical isolates of Enterobacter and Klebsiella species. To avoid further selection and spread of silver-resistant bacteria with a high potential for healthcare-associated infections, the use of silver-based products needs to be controlled and the silver resistance monitored., (Copyright © 2017 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.)

Outbreaks with Enterobacter spp. have been described frequently in neonatal intensive care units (NICUs). This study investigated the factors that determine whether a neonate becomes colonised with Enterobacter spp., how long colonisation continues and whether the termination of isolation measures leads to spread of the organism. Neonates transferred from the NICUs of tertiary care hospitals were screened for the presence of Enterobacter spp. and any potential predictors for colonisation recorded. Those infected were monitored during their hospital stay and colonised neonates were screened every month for six months. Isolation infection control precautions were lifted and all neonates were screened for the presence of Enterobacter spp. six and 12 months later. Fifteen colonised neonates and 33 non-colonised controls were identified for study. Multivariate analysis showed that antibiotic therapy for more than three days and an Apgar score of <8 after 1 min were independently associated with Enterobacter spp. colonisation. Molecular typing using single-enzyme amplified-fragment length polymorphism (seAFLP) analysis revealed 22 different seAFLP genotypes. Three infants remained colonised with the same Enterobacter genotype after discharge; however, most neonates lost their strain or became colonised with another genotype. Lifting infection control measures for neonates colonised with Enterobacter spp. in a neonatal ward did not lead to increased incidence of colonisation and none of the infants became infected. Isolating neonates with susceptible Enterobacter spp. was not found to be necessary.

Comparisons were made between the O-serotype, API 20E profile, site of isolation and antimicrobial resistance of clinical isolates of Enterobacter cloacae. Correlations were found between autoagglutinable strains and urinary-tract infection, and API 20E profile 3305573 and strains isolated from blood. The proportion of strains sensitive to amikacin, gentamicin, cefotaxime, cefuroxime and trimethoprim were 100%, 93%, 91%, 83% and 89%, respectively. No individual resistances or patterns of resistance were associated with O-serotype or biochemical profile. Strains isolated from urinary-tract infections were the most resistant, 40% being resistant to five or more antimicrobials compared to 18%, 12% and 4% for strains from blood, wounds and sputum, respectively. There were no readily identifiable phenotypes within E. cloacae that possessed unique characteristics that could contribute to infections in hospitals.

The observation that Enterobacter cloacae has emerged as a cause of infection in a haematology/oncology ward prompted a prospective study over an eight-month period. Sensitivity to antibiotics, plasmid screening and typing methods were used to describe the epidemiology of the organism which was isolated from 22 patients. There was evidence of a limited amount of cross infection. No common sources were found. A large number of distinct strains were identified. E. cloacae infection was associated with prolonged profound leucopenia and broad-spectrum antibiotic therapy.